China
Africa
Explore chapters and articles related to this topic
AI Is One Thing and Many Things
Published in Tom Lawry, AI in Health, 2020
Alan Turing was an English mathematician, computer scientist and theoretical biologist who is considered to be the father of modern computer science. He formed the concept of algorithms and came up with the idea of a machine that was capable of computing anything that could be computed. His seminal work is the basis for the 2014 movie The Imitation Game. In this film, Turing is seen using an early form of computer science to crack the secret code of the Germans. His work saved lives and helped to end World War II.
Techno-optimism and optimization in media architecture practice and theory
Published in Digital Creativity, 2023
Topic modelling or topic detection, a machine learning method to discover topics in text, can be applied to text on different levels: a corpus, a book, a paragraph. An important question when working with social media discourse concerns the way information emerges from textual data. Machine learning algorithms that facilitate the emergence of unlabelled classes and their organization are of particular interest to this analysis. The self-organizing map (SOM) is a machine learning algorithm introduced by the Finish computer scientist Kohonen (1982), for ordering high-dimensional statistical data so that alike inputs are mapped closer to each other, illustrating the similarity relationships between different data items (such as text documents) in a familiar and intuitive manner. Researchers have investigated the use of SOM for topic modelling and sentiment analysis. Kohonen himself published extensively on his work with text (Honkela et al. 1996; Ritter and Kohonen 1989; Kohonen 1998). He described how entire documents can be distinguished from each other based on their statistical models. Lee and Yang (1999) created word and document cluster maps using SOM to cluster words of similar meaning and documents with overlapping words (similar discourse), on a corpus of news from a Chinese News Site. Using SOM to facilitate analytical inquiries into relationships of words and documents, they demonstrated the power of unsupervised clustering without pre-existent categories.
Applying machine learning to market analysis: Knowing your luxury consumer
Published in Journal of Management Analytics, 2019
Kuo Chi-Hsien, Shinya Nagasawa
Random Forest is a model mainly based on decision-tree (Liaw & Wiener, 2002) (See Figure 3). As mentioned above, the decision tree is a rule-based model, which uses the attributes with the highest covariance as its root, handling the most crucial data first. At this point, if each decision tree is a domain expert, each expert will have their preference. However, dealing with the most informative attribute first may not always be the correct approach. Some minor attributes may contain information strongly related to the desired result. Under these circumstances, computer scientist Tin Kam Ho at IBM research proposed the random forest algorithm (Ho, 1995). In this machine-learning algorithm, the random forest model will create lots of different decision trees, and randomly choose the attributes to process first. The probability for each attribute should be the same. Then, during each training phase, the decision tree algorithm will optimize its result based on the training data, and the random forest model will collect these trees to build a forest. In the testing phase, when new data comes in, the random forest model will consult each decision tree to ask their opinions. In the end, the random forest model will collect opinions from all trees, and take a majority vote to report its answer. Unlike the decision tree, which has a strong bias to the most influential factors, the random forest has proven to have less bias, and generally better accuracy (Geurts et al., 2006).
A comparative, sociotechnical design perspective on Responsible Innovation: multidisciplinary research and education on digitized energy and Automated Vehicles
Published in Journal of Responsible Innovation, 2021
David J. Hess, Dasom Lee, Bianca Biebl, Martin Fränzle, Sebastian Lehnhoff, Himanshu Neema, Jürgen Niehaus, Alexander Pretschner, Janos Sztipanovits
The most developed integration of social sciences and computer scientists in the research involved two social scientists (one faculty, one graduate student), a computer scientist (faculty), and undergraduate computer science students.